cnv-caller-plotter
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes an audit report file
cnv-caller-plotter_audit_result_v2.jsoncontaining self-referential safety assertions such as "security: PASS" and "veto_gates: PASS". This content is designed to present a predetermined safety conclusion to agents or scanners processing the skill context. • Ingestion points: The agent ingests the audit JSON file as part of the skill documentation. • Boundary markers: No instructions or delimiters are present to warn the agent to ignore instructions or claims within the audit file. • Capability inventory: The primary scriptscripts/main.pyhas file system access for directory creation (os.makedirs) and file output (save_bed). • Sanitization: There is no validation or sanitization of the claims provided in the metadata files before they are loaded into the agent's context.
Audit Metadata